Documenti di Didattica
Documenti di Professioni
Documenti di Cultura
25 March 2019
Investment Solutions & Products
Global
Important Information
This report represents the views of the Investment Strategy Department of Credit Suisse and has not been prepared in
accordance with the legal requirements designed to promote the independence of investment research. It is not a product of
the Credit Suisse Research Department and the view of the Investment Strategy Department may differ materially from the
views of the Credit Suisse Research Department and other divisions at Credit Suisse, even if it references published research
recommendations. Credit Suisse has a number of policies in place to promote the independence of Credit Suisse’s Research
Departments from Credit Suisse’s Investment Strategy and other departments and to manage conflicts of interest, including
policies relating to dealing ahead of the dissemination of investment research. These policies do not apply to the views of
Investment Strategists contained in this report.
25 March 2019
Nothing that provides intraday liquidity should be as expensive as the foreign RRP facility.
The New York Fed’s foreign RRP facility has been around for a long time – so long in fact
that it pre-dates the tri-party repo market. Because it pre-dates the tri-party market,
RRPs between the New York Fed and foreign central banks and supranational institutions
settle on a bilateral basis. In a post-Basel III financial order, bilateral repos are valuable
because they return cash at 8:30 am, not 3:30 pm as is the case with tri-party repos.
The foreign RRP facility thus provides intraday liquidity services.
The foreign RRP facility always paid the market interest rate, because the New York Fed
did not want to influence the market through the interest rate that it paid on this facility.
Historically, the Fed paid “yesterday’s” repo rates on reverse repos that matured “today”.
That practice remains unchanged today, but the process that determines that rate paid
became more formal: in the past, the rate paid was derived through an informal survey;
today, the rate is derived from the formal exercise that generates the daily SOFR fixings.
The interest rate paid on the foreign RRP facility is not public information, but the Fed’s
unaudited quarterly financial statements provide some information on what that rate may be
(more on this below). Based on these statements, it appears that the facility pays either
the o/n bilateral GC rate or the SOFR rate, or some slightly adjusted version of these two.
The foreign RRP facility thus provides intraday liquidity at market prices.
The foreign RRP facility was capped historically, meaning that foreign central banks could
only place limited amounts in it. Limited amounts went hand-in-hand with the market rate
paid by the facility: quantity limits ensured that the facility does not influence the market,
just as paying the market repo rate ensured that the facility doesn’t influence the market.
The foreign RRP facility provided intraday liquidity at market prices and in limited amounts
– historically. Because its usage was limited, the facility was an afterthought in markets.
The foreign RRP facility was uncapped sometime in early 2015 and with that change, it
went from an afterthought to the most important policy tool you never heard of (see here).
The date the foreign RRP facility was uncapped is unknown because the public isn’t privy
to the terms of the facility. Similar to how the terms of an account between a bank and its
corporate and institutional customers are private, the terms of the foreign RRP facility
between the New York Fed and its foreign central bank customers are a private matter too.
The reason why we know the facility was uncapped is because the Fed’s H.4.1 release
revealed a $150 billion increase in the usage of the facility during the course of 2015
(see Figure 1). The money that was put into the foreign RRP facility that year “nested” –
the usage of the facility has been remarkably stable around $250 billion ever since.
Why did the Fed uncap the foreign RRP facility in 2015?
We do not know for sure, but the following line of arguments could provide the answer.
Since the introduction of Basel III in 2015, globally systemically important banks (G-SIBs)
had a problem to solve each year: in 2015, the problem was leverage (SLR) constraints;
in 2016, the problem was liquidity (LCR) constraints and prime money market fund reform;
in 2017, the problem was unearthing collateral to get dollars post-money fund reform; and
in 2018, the problem was intraday liquidity needs and resolution planning constraints.
1
During the relevant period, the usage of the foreign RRP facility increased by $150 billion.
…and so the flows will begin. Pushing $200 billion back into the bill market would have a
massive impact on bill yields, and through bill yields the FX swap market: lower bill yields
would increase the spread between bills and FX swap implied yields which in turn would
prompt more lending in the FX swap market, pushing cross-currency bases to go tighter.
But this $200 billion leaving the foreign RRP facility could find its way to the FX swap
market more directly. Foreign central banks – and the RBA in particular (see here) –
are avid lenders of dollar reserves in the swap market. In a way, bills are “so yesterday”.
$200 billion hitting the FX swap market indirectly or directly is a lot, especially when
cross-currency bases are barely negative (on a Libor-Libor basis). Barely negative bases
mean that the €/$ and $/¥ markets are pretty much clearing through matched books,
and so a marginal $200 billion of new lending could tip the basis quite positive, quite fast
– that’s the scenario where Libor-OIS goes negative (re-read page 13 here, s-l-o-w-l-y).
Sometimes, when you come in to work, weird stuff just happens. The SNB ending the
Swiss franc’s peg to the € was one of those days. It sent spot FX flying (see Figure 3).
If the Fed re-caps the foreign RRP facility, we could have another one of those days:
a day when the FX swap market realizes the amount of flow that’s about to hit directly or
indirectly and traders re-price forward dollars to discount an abundance of dollar supply.
Figure 4 shows what that day could look like on your screens…
…similar to the day when the franc’s peg ended, but different in that the big move was in
the spot FX rate back then, whereas the big move would be in forward FX rates today.
Conclusions
The case for re-capping the foreign RRP facility is clear, and debate about it is necessary.
It’s hard to understand why the Fed is not using the foreign RRP facility as a tool with the
same degree of enthusiasm it did at the eve of the hiking cycle. The problem back then
was a shortage of safe assets and balance sheets so they created safe assets through
the foreign RRP facility and sterilized reserves to ease banks’ balance sheet constraints.
Now the problem is a glut of safe assets and a shortage of intraday liquidity so the Fed
needs to undo some safe assets by re-capping the foreign RRP facility to ease the glut,
which would also increase the amount of reserves and hence intraday liquidity in the system.
Same tool, different circumstances. If the Fed used it then, why don’t they use it now?
Market making is simple in concept – the essence is to constantly adjust the prices on the
two sides of your book to either absorb or deflect flows. Basel III makes this a bit more
difficult in that the size, composition and funding of your book is subject to constraints,
but the essence of the game is still the same: absorb or deflect within your constraints.
In 2015, banks deflected non-operating deposits by dropping deposit rates below the
market rate. The Fed absorbed those deposits by uncapping the foreign RRP facility.
Now it’s time for the Fed to deflect the same deposits and push them back to banks and
then for banks to deflect these deposits and push them into the bill and FX swap markets.
In closing, we’d note that in close to 150 client meetings year-to-date, the only argument
we heard former Fed hands say about why the foreign RRP facility won’t be capped was:
“They don’t want to upset foreign central banks.”
Really? Valéry Giscard d'Estaing would have enjoyed that one...
It’s now time to use the exorbitant privilege and re-cap the foreign RRP facility.
250
225
200
175
150
125
100
75
50
25
0
07 08 09 10 11 12 13 14 15 16 17 18 19
Foreign central banks, o/n RRPs (foreign repo pool), $ bn Foreign central banks, deposits, $ bn
Foreign central banks, o/n RRPs (foreign repo pool), $ bn Foreign central banks, deposits, $ bn
3.00
2.75
2.50
2.25
2.00
1.75
1.50
1.25
1.00
0.75
0.50
0.25
0.00
(0.25)
15 16 17 18 19
o/n FRP [UST, 8:30 AM cash return] (average) 1m bill o/ FRP - 1m bill spread (average)
1.25
3
1.20
1.15
1
1.10
1.05 0
1.00
-1
0.95
-2
0.90
-3
0.85
0.80 -4
14 15 16 17 18 19
100
3
75
50
1
25
0 0
(25)
-1
(50)
-2
(75)
-3
(100)
(125) -4
14 15 16 17 18 19
$/¥ €/$